public inbox for linux-mtd@lists.infradead.org
 help / color / mirror / Atom feed
* JFFS2_WBUF_VERIFY patch clarification (RESENT)
@ 2011-03-23  5:40 Sankara Narayanan Balasubramanian
  2011-03-31 14:34 ` Artem Bityutskiy
  0 siblings, 1 reply; 2+ messages in thread
From: Sankara Narayanan Balasubramanian @ 2011-03-23  5:40 UTC (permalink / raw)
  To: linux-mtd@lists.infradead.org; +Cc: Somashekar Umadi

Hi,

We are using JFFS2 as part of our linux kernel v2.6.18 in our product. We are using JFFS2 file system on top of the MTD Driver for NAND Flash (SLC Nand flash). We observe the problem during our long duration file write tests (endurance tests) that occasionally the data written through the fwrite function call is not present in the flash. 

We checked the latest kernel sources (2.6.37) and observed a similar problem being resolved through the JFFS2_WBUF_VERIFY patch. We took only that patch (implementation into our wbuf.c file) and we observe that the problem is resolved. We request you to provide the feedback for the following.

1.	Can this patch be taken as a stand-alone patch into our existing JFFS2 as part of linux 2.6.18 release.
2.	What is the exact behavior of the patch? We observe that the patch ensures that the jffs2_wbuf_recovery is called once more if the first attempt to verify fails. How does it ensure that the data is written into the flash properly? 
3.	Are there any chances / known instances that the patch may fail (if the second attempt to recovery also fail)?

Pls note that we cannot move to the latest version of kernel as this product is in maintenance phase.

With regards,
Sankara Narayanan B

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: JFFS2_WBUF_VERIFY patch clarification (RESENT)
  2011-03-23  5:40 JFFS2_WBUF_VERIFY patch clarification (RESENT) Sankara Narayanan Balasubramanian
@ 2011-03-31 14:34 ` Artem Bityutskiy
  0 siblings, 0 replies; 2+ messages in thread
From: Artem Bityutskiy @ 2011-03-31 14:34 UTC (permalink / raw)
  To: Sankara Narayanan Balasubramanian
  Cc: Somashekar Umadi, linux-mtd@lists.infradead.org

On Wed, 2011-03-23 at 05:40 +0000, Sankara Narayanan Balasubramanian
wrote:
> Hi,
> 
> We are using JFFS2 as part of our linux kernel v2.6.18 in our product.
> We are using JFFS2 file system on top of the MTD Driver for NAND Flash
> (SLC Nand flash). We observe the problem during our long duration file
> write tests (endurance tests) that occasionally the data written
> through the fwrite function call is not present in the flash. 
> 
> We checked the latest kernel sources (2.6.37) and observed a similar
> problem being resolved through the JFFS2_WBUF_VERIFY patch. We took
> only that patch (implementation into our wbuf.c file) and we observe
> that the problem is resolved. We request you to provide the feedback
> for the following.
> 
> 1.	Can this patch be taken as a stand-alone patch into our existing
> JFFS2 as part of linux 2.6.18 release

No, this release is not supported by anyone anymore. I think the oldest
long-term stable kernel is 2.6.27.

> 2.	What is the exact behavior of the patch? We observe that the patch
> ensures that the jffs2_wbuf_recovery is called once more if the first
> attempt to verify fails. How does it ensure that the data is written
> into the flash properly? 

Which patch? Specify commit id or send it.

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2011-03-31 14:36 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-03-23  5:40 JFFS2_WBUF_VERIFY patch clarification (RESENT) Sankara Narayanan Balasubramanian
2011-03-31 14:34 ` Artem Bityutskiy

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox